US7994459B2 - Camera-based heliostat calibration with artificial light sources - Google Patents
Camera-based heliostat calibration with artificial light sources Download PDFInfo
- Publication number
- US7994459B2 US7994459B2 US12/562,990 US56299009A US7994459B2 US 7994459 B2 US7994459 B2 US 7994459B2 US 56299009 A US56299009 A US 56299009A US 7994459 B2 US7994459 B2 US 7994459B2
- Authority
- US
- United States
- Prior art keywords
- heliostat
- heliostats
- artificial light
- light source
- cameras
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active, expires
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/78—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using electromagnetic waves other than radio waves
- G01S3/7803—Means for monitoring or calibrating
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24S—SOLAR HEAT COLLECTORS; SOLAR HEAT SYSTEMS
- F24S20/00—Solar heat collectors specially adapted for particular uses or environments
- F24S20/20—Solar heat collectors for receiving concentrated solar energy, e.g. receivers for solar power plants
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24S—SOLAR HEAT COLLECTORS; SOLAR HEAT SYSTEMS
- F24S23/00—Arrangements for concentrating solar-rays for solar heat collectors
- F24S23/70—Arrangements for concentrating solar-rays for solar heat collectors with reflectors
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24S—SOLAR HEAT COLLECTORS; SOLAR HEAT SYSTEMS
- F24S50/00—Arrangements for controlling solar heat collectors
- F24S50/20—Arrangements for controlling solar heat collectors for tracking
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/78—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using electromagnetic waves other than radio waves
- G01S3/782—Systems for determining direction or deviation from predetermined direction
- G01S3/785—Systems for determining direction or deviation from predetermined direction using adjustment of orientation of directivity characteristics of a detector or detector system to give a desired condition of signal derived from that detector or detector system
- G01S3/786—Systems for determining direction or deviation from predetermined direction using adjustment of orientation of directivity characteristics of a detector or detector system to give a desired condition of signal derived from that detector or detector system the desired condition being maintained automatically
- G01S3/7861—Solar tracking systems
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24S—SOLAR HEAT COLLECTORS; SOLAR HEAT SYSTEMS
- F24S50/00—Arrangements for controlling solar heat collectors
- F24S50/20—Arrangements for controlling solar heat collectors for tracking
- F24S2050/25—Calibration means; Methods for initial positioning of solar concentrators or solar receivers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/40—Solar thermal energy, e.g. solar towers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/40—Solar thermal energy, e.g. solar towers
- Y02E10/47—Mountings or tracking
Definitions
- the present invention relates generally to heliostat tracking calibration in a solar power plant, and more particularly to heliostat tracking calibration based on artificial light sources.
- Heliostat calibration refers to estimating a set of parameters, e.g., mounting position and angular pose, which are necessary for issuing tracking efficient commands to a heliostat in a central receiver plant.
- U.S. Pat. No. 4,564,275, issued Jan. 14, 1986 to K. Stone describes a calibration method and system whereby heliostats are commanded to reflect the sun to a calibrating target screen located below the actual receiver aperture. Based in part on an expected time-dependent position of the sun in the sky, an error signal may be defined as the difference between the expected reflected image centroid and the actual centroid of the reflected image on the target, as detected by a ground-mounted camera system.
- the superposition of reflections from multiple heliostats onto the target screen limits or prevents the calibration of multiple mirrors concurrently: that is, only one or very few heliostats can be calibrated at a time.
- the error signals are measured for several Sun positions over the course of several days, after which the heliostat pose or set of tracking parameters that best fits the data is estimated by an optimization process.
- the sun is used as the directional light source. This implies reflection by a heliostat is only detected following a two-dimensional angular search around a starting position, e.g., computed by inverse kinematics using nominal geometric parameters.
- the set of heliostat configurations e.g., in an ordered rotational pair of angles azimuth-elevation (az-el) or an unordered rotational pair of angles (tilt-tilt) space, which produce detectable reflections is relatively small ( ⁇ 10 mrad diameter, corresponding to the angular size of the sun), which greatly delays the calibration process when poor nominal estimates of heliostat parameters are available.
- these systems depend on the sun being present for clear day calibration only, ruling out calibration at cloudy days or nighttime.
- System and method embodiments of the present invention provide for the calibrating of heliostat parameters for subsequent open-loop sun-tracking, the calibration based on driving artificial light reflections from one or more heliostats into one or more image sensors.
- the present invention employs one or more artificial light sources along with a specialized search and parameter estimation procedure.
- Embodiments of the present invention comprise one or more of the following features:
- the invention in its several embodiments comprises a system and method for heliostat calibration.
- exemplary system embodiments may comprise: one or more artificial light sources, such as curvilinear light sources that may comprise lineal light sources, and/or one or more point light sources; set of image sensors such as one or more cameras; a set of heliostats that may be re-oriented via one more axes of control motion; a target such as a tower receiver, and a computing platform comprising computer-executable instructions and/or circuitry that, in cooperation with heliostat actuators, effects the motions of heliostats during calibration and sun-tracking.
- the system may further comprise a surveying data store comprising positional information for each image sensor, e.g.
- Some system embodiments have a controller module configured to command one or more heliostats to search for configurations that produce a reflection of a light source of choice in such a way that it is detectable by a camera of choice, and issuing a record to a sample data store comprising information related to the configuration, e.g., the angular orientation, at which one or more such reflections were produced, as well as the time at which such an event was detected, the sun position, and the heliostat unique identifier.
- Some system embodiments of the present invention may further comprise a sample data store that stores detected reflection information as produced by the controller.
- Some system embodiments of the present invention may further comprise a module configured to estimate for one or more heliostats, using an optimization procedure, a set of heliostat geometric or kinematic parameters which best fits or explains the set of samples gathered for that heliostat by the said controller.
- the controller may be configured to generate heliostat re-orientation commands to direct reflections for calibration and the reflection may be arrived at in one or all of the following methods: (a) generating an inverse kinematic estimation of a starting position; (b) generating a spiral or piecewise linear motion of the heliostat in its configuration space; and/or (c) a centering procedure after an initial reflection is detected which estimates the centroid of a contiguous region of detectable reflections.
- Some system embodiments may comprise one or more light sources that are substantially vertical, with known heights for lower and upper endpoints, and in some embodiments, these vertical light sources may further comprise one or more point-sources which are separately mounted or which are mounted within their spans (termed “internal”), with a feature which is separately detectable, such as a color contrast or flashing. Accordingly, methods of separate detection of reflection of the present invention may be based on a colored main light source stem and differently colored internal point-sources. Also, methods of separate detection may be based on a main light source stem being of a single, constant brightness, and the internal point-sources having a flashing pattern, or vice-versa.
- Some system embodiments of the present invention may comprise one or more curvilinear light sources, that may comprise lineal light sources, and the curvilinear light sources may be oriented substantially horizontal, at an angle with the horizontal, or non-straight.
- Some system embodiments of the present invention comprise one or more vertical curvilinear light sources having one or more perpendicular curvilinear light sources that intersect at various positions along the vertical curvilinear light source.
- the imaging sensors of some system embodiments of the present invention comprise cameras with an aperture which is a pinhole, a lens system, or a wide-angle lens system and the imaging sensors may contain attenuating filtering. To maximize contrast and/or minimize noise, artificial light based calibrations may be conducted after nightfall and before sunrise, or during an eclipse of the sun.
- some system embodiments of the present invention include a sun-tracking system for a central receiver solar power plant, comprising: (a) a heliostat field comprising heliostats for reflecting sunlight to a receiver; (b) a plurality of artificial light sources that may include flashing lights and may include one or more flashing lights comprising at least one low-duty cycle, high intensity, flashing light; (c) one or more cameras directed toward at least a subset of the heliostats and configured for producing images of the heliostats where a camera may be configured with a shutter or gain control permitting synchronized receptions of artificial light, particularly high intensity light flashing at a low duty cycle during daylight or moonlight; and (d) a controller configured for processing the images, controlling the heliostats, and configured for estimating parameters of the heliostats for open-loop sun-tracking based on artificial light source reflections that may include flashing light source reflections and may include low-duty cycle, high intensity light source reflections, from one or more
- the system controller and camera may be configured for operation during at least one of: daylight and moonlight.
- the plurality of artificial light sources comprises one or more flashing lights
- at least one of the one or more cameras directed toward at least a subset of the heliostats may be further configured for synchronized image detection of the one or more flashing lights
- the controller may be further configured for estimating parameters of the heliostats for open-loop sun-tracking based on flashing artificial light source reflections from one or more heliostats, received by the one or more cameras.
- Method embodiments of the present invention include detecting one or more artificial light reflections such as curvilinear light source reflections, and that may be detected by an imaging sensor as a substantially contiguous group of discrete pixels or “blob,” in binary, or multi-bit format.
- Method embodiments of the present invention provide for more than 50 heliostat reflections processed simultaneously and independently at one or more two-dimensional imagers such as cameras.
- a reflection search and centering procedure of some method embodiments of the present invention may be based on a priori, e.g., survey-derived, engineering drawing derived, or otherwise known, geometry of the locus of detectable reflections in the heliostats configuration space, and the configuration space search process may comprise driving a heliostat in a single degree of freedom toward the locus, or in one or more alternate single-degree of freedom motions, or all degrees of freedom simultaneously.
- Some reflection search and centering procedure embodiments explore one or more features of a single light source, such as detecting internal point-sources, detecting endpoints, detecting changes in segmented features; and the parameter estimation is achieved via the minimization of errors with respect to the expected geometry.
- Some method embodiments apply standard Non-Linear Least-Squares optimizer to estimate best-fit calibrated parameters. Some method embodiments of the present invention may collect sufficient samples to overly constrain the solution set, and thereby defining more constraints than there are parameters to be estimated, leading to robustness of results with respect to noise in individual samples. Some method embodiments of the present invention may comprise driving heliostats, having estimated, i.e., calibrated, parameters, in open-loop during sun-tracking based on a sun-positioning algorithm. Some method embodiments of the invention have a small set of tracking heliostats replaced in the list of calibrating heliostats so as to refresh their internal parameters and/or update them with respect to geometric shifts over time.
- Some method embodiments of the present invention apply a criterion based on angular distance and/or linear independence to schedule reflection searching by a heliostat with a camera pair, a curvilinear light source, and one or more criteria based on expected path length may be used to schedule reflection searches into a camera from a set of two or more light sources.
- Some method embodiments of the present invention may be used entirely independently or in conjunction with a sun-based light-source method described in the prior art, namely, simultaneously during a clear day, or before dawn or after dusk, in alternate fashion with a daytime method.
- some method embodiments of the present invention may be based on at least one camera receiving heliostat reflections of artificial light sources and configured for acquiring heliostat pointing samples in at least one of: daylight and moonlight
- the estimating by the computing device may further comprise estimating heliostat tracking parameters of the heliostats for open-loop sun-tracking based on artificial light source reflections from one or more heliostats, received by the at least one camera in at least one of: daylight and moonlight.
- the plurality of artificial light sources may comprise one or more flashing lights whereat least one camera is directed toward at least a subset of the heliostats may be further configured for synchronized image detection of the one or more flashing lights, and the estimating by the computing device may further comprises estimating parameters of the heliostats for open-loop sun-tracking based on flashing artificial light source reflections, from one or more heliostats, received by the at least one camera.
- Some method embodiments may include a flashing artificial light that flashes, at a low-duty cycle, a high intensity light source.
- Some embodiments of the present invention include a controller comprising a processing unit and addressable member; where the processing unit configured to: (a) receive, by the controller, acquired heliostat pointing samples comprising reflected artificial light via a one or more cameras receiving heliostat reflections of one or more artificial light sources; and
- Embodiments of the present invention include processor readable medium, e.g., installable to the processor from a non-volatile computer-readable medium, having processor executable instructions thereon, which when executed by a processor cause the processor to (not necessarily in the following order): (a) acquire, by a controller, heliostat pointing samples comprising reflected artificial light via one or more cameras receiving heliostat reflections of one or more artificial light sources; (b) estimate heliostat tracking parameters based on the pointing samples; (c) search for a configuration of a first heliostat of the heliostats that reflects light from an artificial light source toward a first camera of the cameras; and (d) store a sample comprising an identifier of the first heliostat, an identifier of the first camera, an indication of the artificial light source directionality, and the heliostat configuration.
- processor executable instructions which when executed by a processor cause the processor to (not necessarily in the following order): (a) acquire, by a controller,
- FIG. 1 is a schematic view of a heliostat reflecting sunlight to a receiver according to embodiments of the present invention
- FIG. 2 is a perspective view of a central-receiver power plant according to embodiments of the present invention prior to the installation of one or more artificial light sources;
- FIG. 3 is a perspective view of a central-receiver power plant according to embodiments of the present invention.
- FIGS. 4A-4D depict vertically-oriented artificial light sources according to embodiments of the present invention.
- FIG. 5 depicts exemplary reflection geometry according to embodiments of the present invention
- FIGS. 6A-6C depict exemplary artificial light reflections viewed on the mirror of a heliostat
- FIG. 7A depicts a perspective view of a heliostat field showing a set of heliostats and reflections of curvilinear light sources
- FIG. 7B depicts an array of “blobs” produced by received reflections at an image sensor
- FIG. 8A depicts an exemplary reflection region in configuration space corresponding to a curvilinear light source
- FIG. 8B depicts an exemplary region in configuration space corresponding to a point light source
- FIG. 9 depicts an exemplary spiral search in configuration space to locate a point source reflection region
- FIG. 10A depicts a search path in configuration space to a reflection region based on motion along one degree-of-freedom
- FIG. 10B depicts a piecewise elementary search path in configuration space
- FIG. 10C depicts a search path in configuration space to a reflection region based on motion along two degrees-of-freedom
- FIG. 11 depicts a search path in configuration space for a vertical, curvilinear light source having point source light emitters in its interior, i.e., within its span;
- FIG. 12B depicts in configuration space, a search path starting a location S and visiting reflection regions corresponding to the light sources of FIG. 12A ;
- FIG. 14 is a top-level process flowchart depicting steps of calibrating based on a reflected light source.
- FIG. 15 is a top-level process flowchart depicting steps of calibrating based on a reflected curvilinear light source having internal beacons.
- one or more cameras may be mounted atop poles and directed toward the heliostat field.
- a heliostat which reflects the sun toward a camera can be imaged, under suitable filtering, as a small compact region of the image sensor—a “blob”; non-contiguous heliostats can be imaged simultaneously at non-adjacent regions of the image sensor, and therefore significant parallelism is possible, and in practice it may be limited by the heat-rejection system protecting the camera and/or the proximity of far-away heliostats as they are imaged onto the image sensor. Accordingly, heliostat tracking parameters may be estimated after each heliostat's mirror is oriented so as to produce reflections into all cameras visible to it, e.g., in round robin fashion, and over several sun positions.
- Heliostat calibration refers to estimating the geometric parameters necessary for a heliostat to reflect the sun continuously into a receiver according to open-loop control commands. Such estimation is typically required since such parameters are not known to a sufficient level of accuracy, following manufacturing and installation processes. For example, sufficient accuracy in the six rigid body parameters of a heliostat's mirror is required typically for a given concentration to be achieved at a target. Likewise parameters internal to a heliostat kinematics, e.g., as produced by a manufacturing process, might not be repeatable under commercially acceptable cost constraints.
- azimuth-elevation or tilt-tilt kinematic may prescribe that its axes of rotation be exactly perpendicular, when in fact, a certain perpendicularity tolerance (to be estimated) will be associated with each manufactured unit.
- Another toleranced manufacturing parameter may be the mounting pose of the mirror with respect to the mirror supporting structure, for example.
- image sensors require little or no sun-filtering or heat-shielding; parallelism, no longer limited by heat considerations, but by the resolution of the sensor, blob size, namely, the detectability of two contiguous heliostats—e.g., at a receding area of the field—as disjoint blobs on the image plane; and calibration execution times are greatly shortened.
- FIG. 1 shows aspects of how a heliostat 130 is modeled according to the present invention.
- the heliostat 130 has a mirrored surface 132 that reflects light from the sun 100 to a receiver 110 .
- the receiver 110 is generally mounted atop a tower 120 .
- Sunlight incident on the receiver 110 is captured by the receiver 110 and transferred as heat to a heat transfer fluid, such as oil, steam, or molten salt, or converted directly to electricity by high-concentration photovoltaic cells.
- the heat transfer fluid may be used directly in some embodiments or later used in connection with a power block containing a turbine to generate electricity.
- the mirrored surface 132 or mirror of the heliostat 130 is a rigid and substantially planar surface in the embodiment of FIG. 1 . In other embodiments, other shapes may be used, for example, a curved or multi-faceted surface. Reflection of sunlight to the receiver 110 depends on the orientation of mirror 132 .
- the geometry of reflected sunlight is shown by unit vectors ⁇ , ⁇ circumflex over (n) ⁇ , and ⁇ circumflex over (r) ⁇ .
- Vector ⁇ represents the direction from a center point O of the mirrored surface 132 of the heliostat 130 to the sun.
- Vector ⁇ circumflex over (n) ⁇ represents the direction normal to the mirrored surface 132 .
- ⁇ circumflex over (n) ⁇ may represent the normal at the center vertex of a concave or multi-faceted surface.
- Vector ⁇ circumflex over (r) ⁇ represents the direction from the center point O of the mirrored surface 132 to a target point R on the receiver 110 .
- ⁇ circumflex over (n) ⁇ bisects the angle between ⁇ and ⁇ circumflex over (r) ⁇ , sunlight will be reflected to the receiver, and the mirror is said to be on-sun.
- a mirror that remains on-sun while the sun is moving is said to be tracking the sun.
- the mirrored surface 132 of the heliostat 130 is mounted on a pedestal 134 in the embodiment of FIG. 1 .
- the heliostat may be mounted on a frame, pole, or other rigid support structure, ground or non-ground penetrating.
- the heliostat kinematics which determines mirror orientation, is parameterized by two rotations.
- the first rotation, ⁇ is about an axis that is substantially perpendicular to the earth's surface.
- the second rotation, ⁇ is about an axis that is substantially tangential to the earth's surface.
- the first rotation, ⁇ may be termed a pan or azimuth motion.
- the second rotation, ⁇ may be termed a tilt or elevation angle.
- the two rotations are about two mutually perpendicular, substantially horizontal axes, which may offer singularity-free tracking in some circumstances.
- the present invention is not restricted to any particular heliostat kinematic, as long as the kinematic is known, and it enables the mirror normal to span the solid angle required for sun-tracking and reflecting light into one or more cameras.
- the mirror may be driven by stepper motors, which, via step counting, conveniently eliminate the need for position encoders and associated electronics, however they may require a home position sensor, such as a limit switch and a mechanical stop.
- the mirror may be driven by servo motors with position encoders, relative or absolute.
- Forward kinematics maps the two-angle heliostat configuration to the mirror orientation, and consequently, its normal.
- the two axes of rotation are aligned with vertical and horizontal directions, perpendicular, and intersect at the center O of the mirrored surface.
- the forward kinematics is given by a product of rotation matrices about the vertical and horizontal axes.
- Inverse kinematics is the process of obtaining the required angular configuration that produces a desired normal to the mirrored surface.
- the forward kinematics of a generalized version of the heliostat 130 may be modeled by a set of parameters, p 1 , p 2 . . . p N .
- the parameters may include rigid body translational and rotational parameters of the heliostat's base frame relative to the frame used to survey the positions of other components of the power plant. Additionally, the parameters may include parameters internal to the heliostat of significant geometric variance such as axis perpendicularity, translational offsets, step-to-orientation non-linearities, mirror mounting angles, etc.
- FIG. 2 shows a perspective view of a solar power plant according to aspects of the present invention.
- the power plant is spread over an area of land 160 .
- the area of land 160 is a rectangle.
- the area of land 160 may be another shape, for example, a circle, an ellipse, a hexagon, an octagon, or even disjoint areas.
- the power plant shown includes two receivers 110 , 111 near a central area of the land 160 . In other embodiments, the number of receivers may be fewer or greater.
- the power plant also includes heliostats 130 a - d distributed across the land 160 to form a heliostat field.
- FIG. 2 shows four heliostats for illustration purposes; however, an embodiment of the invention will generally have many more heliostats, for example, one hundred to hundreds of thousands.
- the power plant also includes cameras 200 A-H directed toward regions of the heliostat field. Under suitable optics, the cameras 200 A-H produce images that are used by image processing to detect reflections from the heliostats.
- the cameras 200 A-H are typically mounted on towers 210 A- 210 H. To provide sufficient visibility of the heliostats to be imaged, the height of a tower will generally be similar to the height of the receiver 110 . Some of the cameras, for example, cameras 200 E, 200 H are mounted onto the same tower. Cameras 200 A, 200 C, 200 D, 200 E are located near the perimeter of the power plant. Camera 200 B is located outside the perimeter of the power plant. Camera 200 F is located within the heliostat field.
- Cameras 200 G are mounted on a tower that also supports the receiver 110 .
- the cameras are separated by a distance similar to the heights of the supporting towers.
- the cameras may be located in differing manners.
- cameras may be mounted on wires, tethered balloons, telescoping towers, or movable tower structures, e.g., mounted on vehicles, which are typically kept stationary during calibration.
- the system uses geometric information about the solar power plant.
- the desired orientation of a heliostat's mirrored surface may be determined by inverse-kinematics based on the direction to the sun and the direction from the center of the mirrored surface to the receiver, and a set of parameters, p 1 , p 2 , . . . p N .
- the current direction to the sun may be obtained by known methods when the current time and the latitude and longitude on the Earth are known.
- the set of heliostat parameters is known only nominally, therefore the desired orientation may only be computed nominally.
- each of the cameras 200 A-H and receivers 110 , 111 is known in a global coordinate system, for example, a celestial coordinate system, or a terrestrial one established prior to the surveying process.
- the locations may be obtained from standard surveying techniques, for example, by using total stations or GPS equipment.
- An optimization procedure using the heliostat configurations that reflect artificial light to the cameras, the location of the cameras, and the artificial light source directions when the reflections were taken may provide an estimate of the geometric parameters of the heliostat.
- a system utilizes a configuration of cameras directed toward a field of heliostats.
- basic operation of a heliostat calibration and tracking control system includes four phases: (1) sample acquisition, (2) parameter estimation, (3) open-loop tracking, and (4) parameter refresh.
- a small, rotating, set of heliostats currently tracking undergoes the first, second, and third phases to refresh their estimated parameters to account for time variability of parameters, for example, ground shift, easing in of supporting structures, or other time-dependent effects.
- the open-loop tracking phase includes performing inverse kinematics using the estimated parameters along with knowledge of the current sun position and target location, e.g., the surveyed receiver aperture location.
- FIG. 3 The major components of an exemplary system are shown in FIG. 3 , comprising a heliostat field F, within which one or more heliostats H 1 , H 2 , H 3 are installed, close to the ground.
- Each heliostat controls the orientation of a mirror, along in two degrees-of-freedom under some kinematic model, e.g., an angular reference frame such as azimuth-elevation or tilt-tilt.
- a receiver R sits atop a receiver tower T R .
- R's aperture is of some finite area, e.g., as required by the flux spread in a given application.
- a heliostat is tracking the sun if the latter is reflected correctly into R.
- a set of vertical poles T 1 , T 2 , . . . T 6 are installed around the boundary and interior of F. Mounted atop one or more such poles are image sensors, e.g., cameras, C 1 , C 2 , . . . C 4 , and/or curvilinear light sources L 1 , L 2 , L 3 , described below.
- Certain poles e.g., T 6
- other towers, such as T 5 may only support a camera, C 4
- Other towers such as T 4 may contain both an artificial light source such as a curvilinear light source, L 3 , and an image sensor, or camera, C 3 .
- a coordinate system is defined at some suitable location.
- an ⁇ x, y, z ⁇ system may be defined along the East, North, and Up directions of a celestial system, as most sun positioning algorithms used during tracking refer to the same system.
- location and internal parameters of each heliostat are known nominally, e.g., from an architectural or engineering blueprint, and may be further refined via embodiments of the calibration system described herein.
- the positions and geometry of the receiver, cameras, and light sources may be known to within a few centimeters, or in a relative frame, a sufficient degree of angular accuracy is present in the system to support further refinement via artificial light-based calibration.
- FIGS. 4A-4D an exemplary curvilinear light source is shown substantially vertical and straight, and is an omni-directional light emitter of a single color, L, that may have a length, i.e., a vertical height, of 2 m to 50 m of length.
- FIG. 4B shows an alternate embodiment in which a continuous vertical light source is emulated by several point sources, B, e.g., an array of light bulbs, linearly juxtaposed.
- B point sources
- FIG. 4C the embodiment of FIG. 4A is repeated with one or more point sources B 1 , B 2 , . . . B N affixed at intermediate points along the (z min ,z max ) interval.
- the height z 2 , z 3 , . . . z n of intermediate point sources is known accurately.
- such internal point sources are separately detectable from L, e.g., by virtue of different colors and/or flashing patterns from L.
- the vertical light source may comprise two or more sections, and in FIG.
- an embodiment of a vertical light source is shown containing two sections L 1 , L 2 , and each separately detectable by an image sensor, e.g., L 1 and L 2 are of different colors, or have a relatively different flashing pattern.
- Other embodiments may make use of a horizontal curvilinear light source mounted perpendicular to a vertical curvilinear light source.
- Other light source embodiments are possible which are non-vertical, namely, horizontal, at an angle with respect to horizontal, curved, and cross-shaped.
- an curvilinear light source may be represented by a parametric form L(t), 0 ⁇ t ⁇ 1, which describes continuously the light-emitting path traversed by it.
- L(t) For the case of a vertical light source of known base location L x , L y and known vertical extent (z min , z max ):
- a single heliostat 530 , H is shown at an intermediate position between a sensor tower T C and a vertical light source tower T L .
- a camera sensor 520 , C i.e. an image sensor at a known location, is shown mounted atop T C , while a vertical light source 510 , L, is affixed to T L .
- H represents the center of its mirror.
- unit vectors ⁇ circumflex over (n) ⁇ , ⁇ circumflex over (l) ⁇ (t), ⁇ represent the mirror normal, the direction to a point on the light source, and the (fixed) direction to the camera sensor 520 , relative to H.
- the configuration space of the heliostat's mirror normal vector may be parameterized by two angles, ( ⁇ , ⁇ ).
- the heliostat may be slewed within a contiguous region which retains the visibility of a portion of L, e.g., as shown in FIG. 6B .
- a center-ray search procedure may be used to estimate the centroid for the set of configurations under which such a reflection is visible.
- FIG. 6C shows a light source such as the exemplary light source of FIG. 4C is detected at a configuration for which both L and the point source light B 1 may be simultaneously detected, e.g. based on a color difference and/or a flashing difference.
- FIG. 7A is the projection of six heliostats in a field 700 , more particularly the projected reflected portions of one or more artificial light sources, in this example curvilinear light sources, toward C, and the projected reflecting array is shown sufficiently separate so the actual light being reflected may be determined uniquely based on heliostat configuration and location.
- FIG. 7B shows how such a composite perspective of FIG.
- each reflecting heliostat is imaged as a separate group of contiguous binary pixels or “blobs.”
- a heliostat's reflection into a camera contains faint thermal energy when compared to the reflection of the sun during sun tracking, the limit of heliostats separately detectable by a single imager is prescribed by its resolution and the maximum number of heliostats selectable in the field which are resolvable at the image plane. Accordingly, the faint thermal energy typically requires imaged heliostats to be non-adjacent to satisfy the exploitable heliostat uniqueness in the calibration processing.
- Embodiments of the present invention comprise a scheduler portion to a controller which works to ensure that heliostats adjacent in the field are not commanded to search for reflective configurations simultaneously toward a single camera.
- a reflection region also known as a “streak,” refers to the characterization of a reflection of a curvilinear light source in the configuration space, ( ⁇ , ⁇ ), of a mirror.
- forward kinematics maps configuration variables, i.e., two angles in the present example, to a unit mirror normal, ⁇ circumflex over (n) ⁇ .
- an azimuth-elevation kinematic produces the following forward map:
- n ⁇ ( Sin ⁇ ( ⁇ ) ⁇ Sin ⁇ ( ⁇ ) Cos ⁇ ( ⁇ ) ⁇ Sin ⁇ ( ⁇ ) Cos ⁇ ( ⁇ ) ) [ Eqn . ⁇ 2 ]
- ⁇ and ⁇ circumflex over (l) ⁇ (t) be defined as above as unit vectors representative of the direction of the camera and light source respectively, the bisection constraint becomes:
- inverse kinematics refers to mapping the required normal to a ( ⁇ , ⁇ ) pair, and in the case of an Azimuth-Elevation (az-el) kinematics, such an inverse map may be expressed as:
- FIG. 8B shows the case where a disc source such as the sun ⁇ is used as the directional light source, producing a point-like region ⁇ ′, the point-like region being thickened by sun-spread and the mirror's finite size, of correct bisection, satisfying:
- the configuration S required for a directional area light source, such as the sun, to be reflected into a known camera aperture, C may be computed from Eqn. 4 and inverse kinematics, using nominal positional and geometric parameters for H, and C, where, if the disc light source is indeed the sun, its position may be obtained from an algorithm known to those of ordinary skill in the art. Due to tolerances in these parameters, S may not, in general, fall inside the region ⁇ ′ of detectable reflections. Given known tolerances in manufacturing and installation, one may define the expected error distance from S to the centroid as may be represented by ⁇ (see FIG. 9 ).
- the particular heliostat may be commanded to execute a spiral-like motion in configuration space, as shown in FIG. 9 .
- the search path may start with departing from S, and then P′ may represent a piecewise linear, or curved, spiraling path that may be executed until a reflection is detected, i.e., when any point in ⁇ ′ is crossed. Because the spiral search spans a substantially two-dimensional region, the time until detection is proportional to ⁇ 2 .
- a second procedure spans the set of reflection-producing configurations, e.g., as produced by incident light spread and/or the mirror's finite dimensions, so as to the estimate the centroid of ⁇ ′.
- a point-source sample termed here a “directional” sample (DS), containing the tuple (C, ⁇ circumflex over (l) ⁇ , ⁇ , ⁇ ) is recorded and associated with the heliostat in question, containing the following data: (a) the camera C (or its index in a list) which has detected the reflection; (b) the direction ⁇ circumflex over (l) ⁇ from which the (sun) light is shining from; and (c) the configuration ( ⁇ , ⁇ ) associated with the centroid of ⁇ ′.
- DS point-source sample
- N is the number of parameters to be estimated for heliostat 530 , H, i.e., the correct set of parameters to be estimated is a point in N-dimensional space
- a PSS imposes, with the exception of a few degenerate cases, two independent constraints in this space, i.e., the PSS is associated with an N-2 dimensional surface in that space.
- FIGS. 10A-10C depict an alternative search process embodiment of the present invention where the region of reflection, ⁇ , of an artificial light source particularly that of a curvilinear light source, as shown in a heliostat's configuration space ( ⁇ , ⁇ ), is searched.
- One may continue to define the expected distance from S to the curved region of reflection as represented by ⁇ .
- the controller may execute steps/tests to determine or otherwise select an elementary path, P, from S; particularly a path involving a single degree of freedom.
- S′ a reflection may be detected.
- a centering procedure may be performed to estimate a center configuration in the degree of freedom which causes a reflection, i.e., a type of cross-sectional centroiding may be performed.
- One exemplary method continues the elementary 1d motion until the reflection disappears, and then, based on the center of the reflection, the process generates configurations as an estimate of the centroid.
- This exemplary search may be comprised of one or more 1d scans, depending on the known structure of the light source. Referring to the exemplary embodiment of FIG.
- NDS typically imposes only one constraint in the space of unknown heliostat parameters since it is associated with a set of possible light source directions.
- a piecewise approach path may be chosen comprising single degree-of-freedom (dof) motions P 1 and P 2 , as shown in FIG. 10B .
- Another exemplary embodiment comprises search path P 3 involving both degrees-of-freedom simultaneously, as shown in FIG. 10C .
- the time until detection may be expected to be proportional to ⁇ , rather the ⁇ 2 proportionality in the point-light sources.
- a DS may be acquired upon detecting a reflection and finding its cross-sectional centroid.
- the method described in FIG. 10A may also contain a time-out test, e.g., if the region of reflection ⁇ was not flagged after a given angular range of motion, an error may be issued, and then, after such a timeout, the search direction may be inverted as to emulate the spiral.
- This exemplary alternative embodiment differs from the previous exemplary method of FIG. 10 A in that given the curvilinear geometry of the region of reflection, the average expected time would typically be linear with respect to ⁇ . That is, if the true sun tracking parameters of the heliostat are known with approximate certainty, the curvilinear region of heliostat configurations for which a light source is detectable by an image sensor works to reduce the number of turns of the spiral needed to obtain said reflection.
- FIG. 11 shows a search path approach for the case of a light source depicted in FIG. 4C .
- the heights (respective to ground) of each point source B 1 , B 2 , B 3 , B 4 are known and equal to z min , z 2 , z 3 , z max .
- An NDS may be acquired then.
- ⁇ is followed along one chosen direction, e.g., a direction of increasing ⁇ , along P 2 , until feature B 2 is detected, where such detection may be on the basis of flashing or a different color. Accordingly, the mirror will be imaged as in FIG. 6C .
- this exemplary configuration may be termed configuration ( ⁇ 2 , ⁇ 2 ) and a DS is acquired because the position of B 2 is known and it serves the process as a directional source.
- Sample Stored Constraints 1 Reflection region is (C, ⁇ L x , L y , ⁇ 1 ⁇ , ⁇ S′ , ⁇ S′ ) NDS, 1 intersected at S′ 2 B 2 detected (C, ⁇ L x , L y , z 2 ⁇ , ⁇ 2 , ⁇ 2 ) DS, 2 3 B1 or upper end of (C, ⁇ L x , L y , z max ⁇ , ⁇ 1 , ⁇ 1 ) DS, 2 reflection region detected 4 B3 detected (C, ⁇ L x , L y , z 3 ⁇ , ⁇ 3 , ⁇ 3 ) DS, 2 5 B4 or lower end of (C, ⁇ L x , L y , z min ⁇ , ⁇ 4 , ⁇ 4 ) DS, 2 reflection region detected Note that a total of four DS and one NDS are acquired above, i
- Orthogonality of a set of constraints is related to the angular distance between the light sources in the case of DSs.
- the most orthogonal exemplary samples in the above set of TABLE 1 are the third and fifth events, since they correspond to the endpoints of the light source.
- Increasing the orthogonality may be achieved via the reflections being obtained from light sources separated widely about the heliostat field—by about a characteristic distance such as receiver tower height. Also, increasing the number of samples beyond minimum orthogonality will filter measurement noise via the law of averages.
- a similar search process may be executed with the split-feature light source of FIG. 4D , with the following phases: (a) an elementary approach from S to the region of reflection until a reflection is detected. NDS acquired; (b) based on the recognized segment, choose a direction to sweep the reflection region, e.g., determine a sweep direction base on a color or flashing pattern; (c) sweep until reflection region is no longer detected, e.g., the sweep has reached an end-point, acquire DSs, and reversed sweep direction; (d) sweep light source till change in color or flashing feature is detected, acquire a DS, since the height at which the feature changes is known; and (e) sweep till the endpoint of other segment is reached, and acquire a DS.
- Some embodiments of the invention comprise several objects, one of which may be the estimation of the unknown geometric parameters required for a heliostat, H, to accurately track the sun in open-loop mode.
- p 1 , p 2 , . . . , p n may represent all such N parameters of a heliostat, such as Cartesian position and Euler pose angles (roll, pitch, and yaw), and internal kinematic parameters such as internal angle between axes, mirror mounting angles, etc.
- ⁇ 1 , ⁇ 2 , . . . , ⁇ M may represent the configuration variables, e.g., pan and tilt angles, where the configuration variables may be presumed to be commanded to sufficient precision by an external controller.
- FK(•) for Forward Kinematic
- ⁇ circumflex over (n) ⁇ FK ( p 1 ,p 2 , . . . , p N , ⁇ 1 , ⁇ 2 , . . . , ⁇ M ).
- L j (t), 0 ⁇ t ⁇ 1 is the parametric which gives the location of points on the light source continuously, e.g., as in Eqn 1 for a vertical light source. If the location along the length where the reflection is produced is not known, it is associated with an NDS, i.e., t may be anywhere in the (0,1) range. If such a location is known, e.g., if reflection production location determination is based on feature-extraction and/or endpoint cutoffs, then it is associated with a DS, i.e., the location along the light is represented by L j (t 0 ), for a known t 0 .
- Parameter estimation may be based on: (a) installing one or more image sensors and one or more light sources around the heliostat field; (b) commanding H along its degrees of freedom to search for a configuration which produces a centered reflection of some known light source L i into a known image sensor C j ; (c) when such a reflection occurs, a sample (C j ,L j (t), ⁇ 1 , ⁇ 2 , . . . , ⁇ M ) is added to a list associated with H; and (d) estimate a best fit set of parameters based on a list or set of samples.
- An optimal parameter set p 1 , p 2 , . . . , p N , may be estimated which best fits that system.
- an error functional may be defined for each equation as the (squared) magnitude (or magnitude squared) of left-hand minus right hand sides of Eqn 6.
- an optimal parameter set may be found which minimizes the sum of such errors, i.e., parameter solutions may be based on a non-linear least squares optimizer. Because there are N parameters, the samples must offer at least N constraints for such a system to be solved.
- DS represent the number of directional samples gathered, e.g., samples from point light sources
- NDS the number of non-directional samples, e.g., gathered from curvilinear light sources. Since directional (resp. non-directional) samples impose two (resp. 1) constraints, a necessary condition is 2DS+NDS ⁇ N for the non-linear system to be solvable, i.e., for an exemplary best-fit solution to be found. This condition is sufficient if said constraints produce enough orthogonality, i.e., the Jacobian of the non-linear system of equations is full-rank.
- FIG. 12B depicts an exemplary configuration space for H showing the five reflection regions, i.e., ⁇ 1 , ⁇ 2 , . . . , ⁇ 5 , associated with reflections from each of five light sources, respectively.
- the locations and orientations of the reflection regions vary within the configuration space ( ⁇ , ⁇ ) as the relative geometry of each light respective to H also varies.
- FIG. 12B shows a starting configuration, S, and a search path composed of motions a, b, c, d, e, and f (shown as simplified straight lines), visited by H as it crosses each of the reflection regions, in this case in the order ⁇ 5 , ⁇ 3 , ⁇ 2 , ⁇ 1 , ⁇ 4 , corresponding to centered reflections from light sources L 5 , L 3 , L 2 , L 1 , L 4 , respectively. While such an ordered crossing may vary from process to process, a preferred embodiment has an order of crossing that is selected to minimize the total expected configuration space path length, as shown in FIG. 12B .
- an overconstrained sample set provides more stable solutions for the heliostat pose variables, i.e., 2DS+NDS>N, or a rectangular Jacobian which is full-rank, so noise and random fluctuations in samples gathered may be averaged or otherwise filtered from the ensemble of data.
- FIG. 13 shows the two sides of the present invention as split 1300 , from a data flow perspective: a hardware section and controller-based software section.
- Hardware includes sensors 1310 and heliostats 1320 .
- the artificial light sources e.g., the curvilinear light sources 1330 may be controllable.
- the artificial light sources may change color or produce flashing patterns responsive to controller signals.
- artificial light source positions, and/or the curvilinear light source orientations may be reoriented responsive to controller signals.
- the software, or data-centric side of the architecture comprises: (a) a survey data store 1340 ; (b) a controller 1350 ; and (c) a sample data store 1360 .
- the survey data store 1340 comprises data gathered during a post-installation physical survey of the field, which typically comprises accurate locations, e.g., x,y,z, locations that may be refined via calibration, of all camera sensors, light source locations and/or parameterization equations, and the receiver aperture geometry and/or location (for sun tracking).
- the survey data store 1340 may comprise nominal positions, e.g., positions as they may be available from an engineering drawing or blueprint, and internal parameters of heliostats, e.g., parameters as measured during manufacturing and installation processes, and such parameters may include both mirror center location, internal angle between axes (typically 90 degrees), desired mirror mounting geometry, etc.
- the controller may comprise three functional modules:
- the controller may be embodied via one or more central processing units (CPUs) having addressable memory and/or stores and the processing may be effected across a local bus and/or across a distributed network.
- CPUs central processing units
- a module of the exemplary controller 1350 described herein may be further subdivided and combined with other functions so long as the function and processes described herein may be performed.
- the various modules may also be implemented in hardware, or a combination of hardward and software, e.g., firmware.
- FIG. 14 shows in a top-level process flowchart 1400 the computer-based method of NDS gathering by the scheduler for a given heliostat H.
- a particular camera-light source pair (C i , L j ) is selected 1410 , e.g., a selection that may be based on a criterion of sufficient angular distance (related to orthogonality) between all previously acquired samples.
- a simplified, e.g., 1d, path is chosen 1430 in H's configuration space likely to intersect with the reflection region. H is moved along such path while such a reflection 1440 is not detected by C i .
- a local centering procedure 1450 is executed which estimates the central axis of the reflection region, since the latter is somewhat thickened by the angular extent of the H's mirror.
- a sample 1460 e.g., a NDS
- the process repeats until enough (based on rank, orthogonality, and noise filtering) samples are acquired from various pairings of light-sources with image-sensors, at which point parameter estimation may be performed by the solver 1470 .
- a path may be followed from S′ along reflection-producing configurations until the upper or lower endpoint of the light source were detected, e.g., on the basis of reflection disappearance. In such, a case a DS would be acquired.
- FIG. 15 shows in a top-level process flowchart 1500 a similar sample gathering, for the case of a vertical light source with internal point-source emitters B i , and/or features, e.g., as shown in FIG. 4C and FIG. 4D .
- the heliostat slews from S to S′, but then proceeds to follow the reflection region 1510 till one or more point source emitters are detected on the basis of some feature change 1520 .
- a directional sample 1530 is acquired. Once one or more are visited, the procedure terminates.
- Some calibration embodiments of the present invention operate after sundown and before daybreak.
- Several heliostats may have their incident artificial light directed to a single camera.
- An entire array of heliostats may be calibrated simultaneously at night where preferably there is sufficient computational power to address the increased parallelism over calibrating portions of the array, and preferably where there is a significant number of image sensors to increase both orthogonality and parallelism.
- Some calibration embodiments of the present invention may comprise light emitters within sun spectrum dips such as those emitting in the narrowband of 1900-1930 nm, i.e., the water absorption band, so as to increase signal-to-noise ratio at the detectors with respect to sunlight.
- the one or more cameras may each be configured with gain control and/or a shutter synchronized to flashing light emissions.
- LED-based lights or Low-Pressure Sodium light sources might be flashed at 1/100 to 1/1000 duty cycles and accordingly, at 100 ⁇ to 1000 ⁇ stronger luminous power, enabling certain embodiments to operate daytime.
- embodiments of the present invention may operate during daylight, in darkness, concurrently or alternatively, in the presence of the sun and/or the moon, by observing one or more flashing lights via one or more synchronized cameras, e.g., low-duty cycle high intensity light flashing lights with synchronized cameras.
- An exemplary low-duty cycle may be a flash of 100 msec followed by an off period of 900 msec during daylight conditions and a flash of 200 msec followed by a off periods of 1.8 sec during moonlight conditions.
- the scheduler may exclude sample acquisition from heliostats whose reflected light might include the sun, a star, the moon, or another luminous object source foreign to the calibration intent, unless it is the intent to use the known position of said objects as an explicit light source (e.g., the position of the moon can be obtained via a moon-positioning algorithm).
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Sustainable Development (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Energy (AREA)
- Thermal Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Studio Devices (AREA)
Abstract
Description
-
- a) calibration may be executed at any time, referencing one or more artificial light sources, e.g. curvilinear linear light sources, or point-like light sources, preferably at night, irrespective of weather
- b) image sensors may not require sun-filtering or heat-shielding (parallelism becomes limited by the resolution of the sensor and blob size)
- c) Owing to the curvilinear geometry of light sources, the region in heliostat configuration space at which a reflection is detected is much wider, greatly reducing the search time expected for detection.
- (b) estimate heliostat tracking parameters based on the pointing samples. In some embodiments the processor of the controller is further configured to (not necessarily in the following order):
- (c) search for a configuration of a first heliostat of the heliostats that reflects light from an artificial light source toward a first camera of the cameras; and (d) store a sample comprising an identifier of the first heliostat, an identifier of the first camera, an indication of the artificial light source directionality, and the heliostat configuration.
Letting ĉ and {circumflex over (l)}(t) be defined as above as unit vectors representative of the direction of the camera and light source respectively, the bisection constraint becomes:
For a choice of t in the (0,1) interval, a unique normal {circumflex over (n)}=(nx, ny, nz) is prescribed; inverse kinematics refers to mapping the required normal to a (θ,φ) pair, and in the case of an Azimuth-Elevation (az-el) kinematics, such an inverse map may be expressed as:
Accordingly, the (θ,φ) which satisfy
Searching for a Reflection
TABLE 1 | |||
Type of | |||
Sample, # | |||
of Solve | |||
Event | Description | Sample Stored | |
1 | Reflection region is | (C, {Lx, Ly, −1}, θS′, φS′) | NDS, 1 |
intersected at S′ | |||
2 | B2 detected | (C, {Lx, Ly, z2}, θ2, φ2) | DS, 2 |
3 | B1 or upper end of | (C, {Lx, Ly, zmax}, θ1, φ1) | DS, 2 |
reflection region | |||
detected | |||
4 | B3 detected | (C, {Lx, Ly, z3}, θ3, φ3) | DS, 2 |
5 | B4 or lower end of | (C, {Lx, Ly, zmin}, θ4, φ4) | DS, 2 |
reflection region | |||
detected | |||
Note that a total of four DS and one NDS are acquired above, i.e., nine total constraints are imposed, and that these samples may span only a small region of the heliostat's range of motion in (θ,φ) space. In general, these constraints may not be sufficiently independent, in the sense of constraint surface orthogonality around the optimal solution. Orthogonality of a set of constraints is related to the angular distance between the light sources in the case of DSs. For example, the most orthogonal exemplary samples in the above set of TABLE 1 are the third and fifth events, since they correspond to the endpoints of the light source. Increasing the orthogonality may be achieved via the reflections being obtained from light sources separated widely about the heliostat field—by about a characteristic distance such as receiver tower height. Also, increasing the number of samples beyond minimum orthogonality will filter measurement noise via the law of averages.
{circumflex over (n)}=FK(p 1 ,p 2 , . . . , p N,θ1,θ2, . . . , θM).
are the unit vectors toward the light source and toward the image sensor. With K samples, a system of K generally nonlinear equations such as represented by Eqn. 6 will be defined. An optimal parameter set: p1, p2, . . . , pN, may be estimated which best fits that system. In an exemplary embodiment, an error functional may be defined for each equation as the (squared) magnitude (or magnitude squared) of left-hand minus right hand sides of Eqn 6. In such an embodiment, an optimal parameter set may be found which minimizes the sum of such errors, i.e., parameter solutions may be based on a non-linear least squares optimizer. Because there are N parameters, the samples must offer at least N constraints for such a system to be solved. Let “DS” represent the number of directional samples gathered, e.g., samples from point light sources, and NDS the number of non-directional samples, e.g., gathered from curvilinear light sources. Since directional (resp. non-directional) samples impose two (resp. 1) constraints, a necessary condition is 2DS+NDS≧N for the non-linear system to be solvable, i.e., for an exemplary best-fit solution to be found. This condition is sufficient if said constraints produce enough orthogonality, i.e., the Jacobian of the non-linear system of equations is full-rank.
Solving Base Pose with Multiple Light Sources
-
- (a) A scheduler, or
scheduler module 1370, that, when executing instruction and/or responsive to circuitry, maintains a list of heliostats having parameters that have not yet been estimated. That is, thescheduler 1370 maintains a list of uncalibrated heliostats, i.e., heliostats whose DS+2NDS is below a minimum desirable threshold and/or their mutual orthogonality and noise-filtering capabilities are still deemed insufficient for robust optimization. Thescheduler 1370 may periodically place uncalibrated heliostats into motion and search patterns to produce artificial light source reflections, e.g., curvilinear light source reflections that may be detected by a camera of choice according to the calibrations instructions. The reflection search process may be initiated using inverse kinematics based on nominal heliostat parameters, and light source and image sensor locations. Thescheduler 1370 also ensures heliostats chosen for search may be resolved independently at the image sensor. That is, thescheduler 1370 may select uncalibrated heliostats based on the extent that they may project onto the image sensor as blobs which are resolvable, i.e., occupying non-intersecting pixel clusters. For example, one criterion may be for adjacent heliostats not to be involved in producing samples with the same camera simultaneously, though they may be commanded to produce samples with different cameras. Once enough samples are acquired for a given heliostat, the parameter estimator, orparameter estimator module 1380, executes parameter estimation via an optimization method. As centered reflections are produced, samples are stored within thesample data store 1360. - (b) A
parameter estimator 1380 receives the list of samples gathered for the heliostat and available at thesample data store 1360 as input to an optimization module which finds a set of parameters that may produce best-fits of the data. The calibrated parameters, i.e., the parameters improved or refined based on the original/nominal set may then be stored back into thesurvey data store 1340. - (c) An open-loop tracker, or open-
loop tracker module 1390, may be based on a sun-tracker algorithm and the parameters computed by theparameter estimator 1380, generates commands to one or more actuators so that the heliostat may track the sun with significant accuracy. It may be presumed that during tracking, both the target receiver position and instantaneous sun direction are known, e.g., as queried from a sun-position algorithm and may be an algorithm as described in “Solar Position Algorithms for Solar Radiation Applications,” Ibrahim Reda and Afshin Andreas, National Renewable Energy Laboratory (NREL), 2008. The required angle of the heliostat normal vector may be calculated from these values, and the motion of the heliostat needed to achieve this normal vector is calculated from its pose.
- (a) A scheduler, or
Claims (29)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/562,990 US7994459B2 (en) | 2008-08-06 | 2009-09-18 | Camera-based heliostat calibration with artificial light sources |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13791708P | 2008-08-06 | 2008-08-06 | |
PCT/US2009/053046 WO2010017415A2 (en) | 2008-08-06 | 2009-08-06 | Camera-based heliostat calibration with artificial light sources |
US12/562,990 US7994459B2 (en) | 2008-08-06 | 2009-09-18 | Camera-based heliostat calibration with artificial light sources |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2009/053046 Continuation-In-Part WO2010017415A2 (en) | 2008-08-06 | 2009-08-06 | Camera-based heliostat calibration with artificial light sources |
Publications (2)
Publication Number | Publication Date |
---|---|
US20100031952A1 US20100031952A1 (en) | 2010-02-11 |
US7994459B2 true US7994459B2 (en) | 2011-08-09 |
Family
ID=41664207
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/562,990 Active 2029-12-31 US7994459B2 (en) | 2008-08-06 | 2009-09-18 | Camera-based heliostat calibration with artificial light sources |
Country Status (2)
Country | Link |
---|---|
US (1) | US7994459B2 (en) |
WO (1) | WO2010017415A2 (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090249787A1 (en) * | 2006-11-13 | 2009-10-08 | Deutsches Zentrum Fuer Luft- Und Raumfahrt E. V. | Method for controlling the alignment of a heliostat with respect to a receiver, heliostat device and solar power plant |
US20110000478A1 (en) * | 2009-07-02 | 2011-01-06 | Dan Reznik | Camera-based heliostat tracking controller |
US20120192857A1 (en) * | 2011-01-31 | 2012-08-02 | Google Inc. | Heliostat Assignment in a Multi-Tower Field |
WO2013041548A3 (en) * | 2011-09-22 | 2013-05-23 | Siemens Aktiengesellschaft | Mirror alignment device or solar tower of a solar tower power plant and corresponding method and solar tower power plant |
US20150226461A1 (en) * | 2012-08-07 | 2015-08-13 | Logos Technologies, Llc | Solar energy collection utilizing heliostats |
US9372159B2 (en) | 2014-09-10 | 2016-06-21 | Esolar, Inc. | System and method for detecting heliostat failures using artificial light sources |
US9528724B1 (en) | 2011-06-08 | 2016-12-27 | Solarreserve Technology, Llc | Apparatus and method for configuring heliostat fields |
CN108225552A (en) * | 2017-12-25 | 2018-06-29 | 中国科学院电工研究所 | Tower power station heliostat field optically focused energy-flux density distribution measurement method |
US10228163B2 (en) | 2014-03-27 | 2019-03-12 | Mitsubishi Hitachi Power Systems, Ltd. | Heliostat calibration device and heliostat calibration method |
US11009263B2 (en) * | 2019-02-25 | 2021-05-18 | Karl von Kries | Systems and methods for altering rotation of a solar rotational manufacturing system |
WO2023111252A1 (en) * | 2021-12-17 | 2023-06-22 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Method for measuring heliostats, and method for calibrating heliostats |
Families Citing this family (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009055624A1 (en) | 2007-10-24 | 2009-04-30 | Esolar, Inc. | Calibration and tracking control of heliostats in a central tower receiver solar power plant |
US7834303B2 (en) * | 2008-06-09 | 2010-11-16 | Ahura Energy Concentrating Systems | Multi-element concentrator system |
US20110120448A1 (en) * | 2009-11-25 | 2011-05-26 | Google Inc. | Heliostat control scheme using cameras |
WO2011150373A1 (en) | 2010-05-28 | 2011-12-01 | Black Swan Solar, Inc. | Heliostat repositioning system and method |
US8674280B1 (en) * | 2010-07-19 | 2014-03-18 | Sandia Corporation | Concentration solar power optimization system and method of using same |
ES2382707B1 (en) * | 2010-10-20 | 2013-04-24 | Abengoa Solar New Technologies S.A. | TOWER RECEIVER CONFIGURATION FOR HIGH POWER. |
ES2389354B1 (en) * | 2010-11-18 | 2013-09-06 | Romero Dario Garcia | AUTOMATIC SYSTEM FOR CALIBRATION OF THERMOSOLAR PLANT HELIOSATA DATA TYPE TOWER. |
CA2819243A1 (en) * | 2010-12-03 | 2012-06-07 | Daniel Fukuba | Robotic heliostat calibration system and method |
US8442790B2 (en) | 2010-12-03 | 2013-05-14 | Qbotix, Inc. | Robotic heliostat calibration system and method |
WO2012083383A1 (en) * | 2010-12-22 | 2012-06-28 | Commonwealth Scientific And Industrial Research Organisation | Heliostat calibration and control |
US20120266938A1 (en) * | 2011-04-25 | 2012-10-25 | Aspect Solar Pte Ltd | Solar tracking system and method for concentrated photovoltaic (cpv) systems |
US9494341B2 (en) | 2011-05-27 | 2016-11-15 | Solarcity Corporation | Solar tracking system employing multiple mobile robots |
US10030833B2 (en) * | 2011-06-03 | 2018-07-24 | Osram Sylvania Inc. | Multimode color tunable light source and daylighting system |
US20130021471A1 (en) * | 2011-07-21 | 2013-01-24 | Google Inc. | Reflective Surface Orientating with Multiple View Ports |
CN102393753B (en) * | 2011-09-27 | 2014-08-20 | 浙江中控太阳能技术有限公司 | Solar thermal power generation heliostat positioning system and method based on multiband light sources |
US9482583B1 (en) * | 2011-10-06 | 2016-11-01 | Esolar, Inc. | Automated heliostat reflectivity measurement system |
US9500390B1 (en) * | 2011-11-14 | 2016-11-22 | Esolar, Inc. | Heliostat field power controller for setting a throttle to determine an optimum distribution of energy |
DE102012021585A1 (en) * | 2012-10-23 | 2014-03-20 | Astrium Gmbh | Method for wireless energy transmission in enclosed spaces |
US20140174430A1 (en) * | 2012-11-13 | 2014-06-26 | Brian M. Fitzgerald | Pyramid lamp medallion control for solar thermal power generation system |
WO2015173899A1 (en) * | 2014-05-14 | 2015-11-19 | 株式会社 日立ハイテクノロジーズ | Solar thermal power generation system, and calibration system for solar thermal power generation system |
US20160025591A1 (en) * | 2014-07-22 | 2016-01-28 | Esolar Inc. | Automated deflectometry system for assessing reflector quality |
AU2016280892A1 (en) * | 2015-06-19 | 2018-01-25 | Solarreserve Technology, Llc | Heliostat characterization using starlight |
DE102015217086B4 (en) * | 2015-09-07 | 2019-11-07 | CSP Services GmbH | Method for measuring heliostats |
ES2607710B1 (en) * | 2015-10-02 | 2017-10-11 | Fundación Cener-Ciemat | Calibration method for heliostats |
CN105242692A (en) * | 2015-10-26 | 2016-01-13 | 苏州华安普电力科技股份有限公司 | Distributed roof photovoltaic generating intelligent monitoring device |
CN108917205B (en) | 2017-03-29 | 2020-09-11 | 考克利尔维修工程有限责任公司 | System and method for thermal-mechanical monitoring of solar receivers |
CN109308078B (en) * | 2017-07-27 | 2021-08-27 | 龙岩智康太阳能科技有限公司 | Heliostat control method and device, computer-readable storage medium and terminal equipment |
CN107678448B (en) * | 2017-11-27 | 2023-06-02 | 上海晶电新能源有限公司 | Sun tracking correction system and method based on celestial body image |
ES2738907B2 (en) * | 2018-07-25 | 2020-12-30 | Fund Cener Ciemat | PROCEDURE FOR CHARACTERIZING REFLECTIVE ELEMENTS FROM THE BEAMS OF LIGHT REFLECTED THROUGH THEM |
US11250587B2 (en) * | 2019-09-27 | 2022-02-15 | Alliance For Sustainable Energy, Llc | Heliostat error detection |
CN112765846B (en) * | 2021-01-04 | 2022-10-25 | 山东电力建设第三工程有限公司 | Evaluation method for heliostat field polymerization performance of tower type solar thermal power plant |
CN112903629B (en) * | 2021-01-15 | 2022-09-09 | 中国计量大学 | A heliostat reflectivity detection method |
CN112945213B (en) * | 2021-02-01 | 2022-12-30 | 上海晶电新能源有限公司 | Angle-based heliostat space positioning system and method |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS55153903A (en) | 1979-05-18 | 1980-12-01 | Agency Of Ind Science & Technol | Angle adjusting method for plane mirror in reflecting mirror for heliostat |
US4564275A (en) | 1984-06-21 | 1986-01-14 | Mcdonnell Douglas Corporation | Automatic heliostat track alignment method |
US5862799A (en) | 1995-06-22 | 1999-01-26 | Yeda Research And Development Company Ltd. | Control of a heliostat field in a solar energy plant |
US20040086021A1 (en) * | 2002-11-01 | 2004-05-06 | Litwin Robert Zachary | Infrared temperature sensors for solar panel |
US20080236568A1 (en) * | 2007-03-30 | 2008-10-02 | Kevin Hickerson | Heliostat with integrated image-based tracking controller |
US20090107485A1 (en) * | 2007-10-24 | 2009-04-30 | Reznik Dan S | Calibration and tracking control of heliostats in a central tower receiver solar power plant |
US20100191378A1 (en) * | 2007-03-26 | 2010-07-29 | Brightsource Industries (Israel) Ltd. | Distributed power towers with differentiated functionalities |
US20110000478A1 (en) * | 2009-07-02 | 2011-01-06 | Dan Reznik | Camera-based heliostat tracking controller |
-
2009
- 2009-08-06 WO PCT/US2009/053046 patent/WO2010017415A2/en active Application Filing
- 2009-09-18 US US12/562,990 patent/US7994459B2/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS55153903A (en) | 1979-05-18 | 1980-12-01 | Agency Of Ind Science & Technol | Angle adjusting method for plane mirror in reflecting mirror for heliostat |
US4564275A (en) | 1984-06-21 | 1986-01-14 | Mcdonnell Douglas Corporation | Automatic heliostat track alignment method |
US5862799A (en) | 1995-06-22 | 1999-01-26 | Yeda Research And Development Company Ltd. | Control of a heliostat field in a solar energy plant |
US20040086021A1 (en) * | 2002-11-01 | 2004-05-06 | Litwin Robert Zachary | Infrared temperature sensors for solar panel |
US6926440B2 (en) * | 2002-11-01 | 2005-08-09 | The Boeing Company | Infrared temperature sensors for solar panel |
US20100191378A1 (en) * | 2007-03-26 | 2010-07-29 | Brightsource Industries (Israel) Ltd. | Distributed power towers with differentiated functionalities |
US20080236568A1 (en) * | 2007-03-30 | 2008-10-02 | Kevin Hickerson | Heliostat with integrated image-based tracking controller |
US7906750B2 (en) * | 2007-03-30 | 2011-03-15 | Esolar, Inc. | Heliostat with integrated image-based tracking controller |
US20090107485A1 (en) * | 2007-10-24 | 2009-04-30 | Reznik Dan S | Calibration and tracking control of heliostats in a central tower receiver solar power plant |
US20110000478A1 (en) * | 2009-07-02 | 2011-01-06 | Dan Reznik | Camera-based heliostat tracking controller |
Non-Patent Citations (1)
Title |
---|
International Search Report for PCT/US09/053046 dated Mar. 12, 2010. |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8651100B2 (en) * | 2006-11-13 | 2014-02-18 | Deutsches Zentrum Fuer Luft- Und Raumfahrt E.V. | Method for controlling the alignment of a heliostat with respect to a receiver, heliostat device and solar power plant |
US20090249787A1 (en) * | 2006-11-13 | 2009-10-08 | Deutsches Zentrum Fuer Luft- Und Raumfahrt E. V. | Method for controlling the alignment of a heliostat with respect to a receiver, heliostat device and solar power plant |
US20110000478A1 (en) * | 2009-07-02 | 2011-01-06 | Dan Reznik | Camera-based heliostat tracking controller |
US20120192857A1 (en) * | 2011-01-31 | 2012-08-02 | Google Inc. | Heliostat Assignment in a Multi-Tower Field |
US9528724B1 (en) | 2011-06-08 | 2016-12-27 | Solarreserve Technology, Llc | Apparatus and method for configuring heliostat fields |
WO2013041548A3 (en) * | 2011-09-22 | 2013-05-23 | Siemens Aktiengesellschaft | Mirror alignment device or solar tower of a solar tower power plant and corresponding method and solar tower power plant |
US20150226461A1 (en) * | 2012-08-07 | 2015-08-13 | Logos Technologies, Llc | Solar energy collection utilizing heliostats |
US10228163B2 (en) | 2014-03-27 | 2019-03-12 | Mitsubishi Hitachi Power Systems, Ltd. | Heliostat calibration device and heliostat calibration method |
US9372159B2 (en) | 2014-09-10 | 2016-06-21 | Esolar, Inc. | System and method for detecting heliostat failures using artificial light sources |
CN108225552A (en) * | 2017-12-25 | 2018-06-29 | 中国科学院电工研究所 | Tower power station heliostat field optically focused energy-flux density distribution measurement method |
CN108225552B (en) * | 2017-12-25 | 2020-04-28 | 中国科学院电工研究所 | Method for measuring concentration energy flux density distribution of heliostat field in tower-type power station |
US11009263B2 (en) * | 2019-02-25 | 2021-05-18 | Karl von Kries | Systems and methods for altering rotation of a solar rotational manufacturing system |
WO2023111252A1 (en) * | 2021-12-17 | 2023-06-22 | Deutsches Zentrum für Luft- und Raumfahrt e.V. | Method for measuring heliostats, and method for calibrating heliostats |
Also Published As
Publication number | Publication date |
---|---|
WO2010017415A2 (en) | 2010-02-11 |
WO2010017415A3 (en) | 2010-05-06 |
US20100031952A1 (en) | 2010-02-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7994459B2 (en) | Camera-based heliostat calibration with artificial light sources | |
US8104893B2 (en) | Calibration and tracking control of heliostats in a central tower receiver solar power plant | |
US8386179B2 (en) | Solar access measurement device | |
US20130021471A1 (en) | Reflective Surface Orientating with Multiple View Ports | |
CN102354225B (en) | Calibration system and calibration method for heliostat of solar generating station | |
US4211922A (en) | Heliostat guidance | |
US20110155119A1 (en) | Heliostat with integrated image-based tracking controller | |
AU2016329628B2 (en) | Calibration method for heliostats | |
CN103874891B (en) | For sensing and point to the optics alternative of aiming light source | |
US20110317876A1 (en) | Optical Control System for Heliostats | |
CN102354224B (en) | Correcting system and method of sunlight reflecting device based on artificial light source | |
CN109508043B (en) | Image-based heliostat secondary reflection pointing correction field system and method | |
ES2617569B2 (en) | HELIOSTAT CALIBRATION DEVICE AND HELIOOSTAT CALIBRATION METHOD | |
EP2534431A2 (en) | Scalable and rapidly deployable master-slave method and apparatus for distributed tracking solar collector and other applications | |
ES2656066T3 (en) | Cosmic radiation concentrator installation equipped with a reflective optical surface control system | |
CN110209205A (en) | A kind of heliostat bearing calibration based on mirror surface label | |
KR102064347B1 (en) | Natural light device and initializing method for the natural light device | |
WO2015173899A1 (en) | Solar thermal power generation system, and calibration system for solar thermal power generation system | |
Burisch et al. | Scalable heliostat calibration system (SHORT)-Calibrate a whole heliostat field in a single night | |
CN118401791A (en) | Method for measuring heliostat and method for calibrating heliostat | |
Aiba et al. | Improvement of image processing control accuracy for cross linear heliostat | |
US11262103B1 (en) | Heliostat localization in camera field-of-view with induced motion | |
Kalele | Improved Algorithm for Solar Power Systems for Efficient Tracking System | |
CN116124179A (en) | Heliostat control system image acquisition module online calibration method | |
CN117029786A (en) | Heliostat installation deviation measurement method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: ESOLAR, INC.,CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ZAVODNY, MAXIMILIAN;REZNIK, DAN S.;FORTE, MICHAEL JAMES;SIGNING DATES FROM 20090803 TO 20090804;REEL/FRAME:023256/0032 Owner name: ESOLAR, INC., CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ZAVODNY, MAXIMILIAN;REZNIK, DAN S.;FORTE, MICHAEL JAMES;SIGNING DATES FROM 20090803 TO 20090804;REEL/FRAME:023256/0032 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
REMI | Maintenance fee reminder mailed | ||
FPAY | Fee payment |
Year of fee payment: 4 |
|
SULP | Surcharge for late payment | ||
AS | Assignment |
Owner name: AALBORG CSP, DENMARK Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:ESOLAR, INC.;REEL/FRAME:046665/0965 Effective date: 20180503 |
|
FEPP | Fee payment procedure |
Free format text: ENTITY STATUS SET TO SMALL (ORIGINAL EVENT CODE: SMAL) |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2552); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY Year of fee payment: 8 |
|
AS | Assignment |
Owner name: SEPCOIII ELECTRIC POWER CONSTRUCTION CO. LTD., CHINA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:AALBORG CSP A/S;REEL/FRAME:051706/0028 Effective date: 20191128 |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 12TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2553); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY Year of fee payment: 12 |